Over half of pregnancies in women with lupus are complicated by miscarriage, preterm birth, or preeclampsia, most commonly in women with significant lupus activity. The guiding principle of lupus management is thus to use the safest available medications to prevent flares during pregnancy. Data supporting this principle are scarce, however, and based on small cohorts. Limitations in the data have also prevented identification of the best treatment strategies for women with varying risks. The goal of this proposal is to identify strategies that optimize pregnancy outcomes using comparative effectiveness approaches. The first method will be a patient- level meta-analysis based on datasets from published lupus pregnancy cohorts; 8 of the largest cohorts have already demonstrated interest. By stratifying pregnancies by lupus activity, prior renal disease, and maternal race, this analysis will yield new data about optimal medication therapies to control lupus and improve pregnancy outcomes. The second method, a decision analysis, will assess potential cost and pregnancy benefits of optimal lupus management. These two aims will offer new data to guide lupus management. Integrating this knowledge into daily practice, however, will be a challenge. For example, current best practices, like continuing hydroxychloroquine (HCQ) in lupus pregnancy, have not yet moved from expert clinics to daily practice. An exploratory aim will use mixed methods to ask community rheumatologists to identify strategies and tools that would enable them to integrate expert recommendations into daily practice. An essential goal is to develop an active Patient Advisory Council (PAC), comprising primarily mothers with lupus, to guide Dr. Clowse in these and future projects. This PAC will be integrated into many aspects of this proposal. Dr. Clowse, a national leader in clinical management of rheumatologic disease in pregnancy, seeks to build new methodology skills to move her research forward. To this end, this proposal outlines a program of research projects, apprenticeships, mentoring, and course work in statistics, comparative effectiveness, and patient-centered outcomes methods. With the primary mentorship of Gillian Sanders-Schmidler, PhD, Director of Duke University's AHRQ-funded Evidence-Based Practice Center, Dr. Clowse will build strong meta- analysis skills. She will practice decision analysis with co-mentor Evan Myers, MD, MPH, lead author for numerous women's health decision analyses and guidelines for AHRQ. At project completion, Dr. Clowse will have the skills, the data, and the stakeholder partnerships needed to drive innovation in the management of lupus in pregnancy.